3D Reconstruction of Asphalt Pavement Macro-Texture Based on Convolutional Neural Network and Monocular Image Depth Estimation
Source abstract
Abstract
The 3D reconstruction of asphalt pavement macrotexture holds significant engineering value for pavement quality assessment and performance monitoring. However, conventional 3D reconstruction methods face challenges, such as high equipment costs and operational complexity, limiting their widespread application in engineering practice. Meanwhile, current deep learning-based monocular image reconstruction for pavement texture remains in its early stages. To address these technical limitations, this study systematically prepared four types of asphalt mixture specimens (AC, SMA, OGFC, and PA) with a total of 14 gradations. High-precision equipment was used to simultaneously capture 2D RGB images and 3D RGB-D point cloud data of the surface texture. An innovative multi-scale feature fusion CNN model was developed based on an encoder–decoder architecture, along with an optimized training strategy for model parameters. For performance evaluation, multiple metrics were employed, including root mean square error (RMSE = 0.491), relative error (REL = 0.102), and accuracy at different thresholds (δ = 1/2/3: 0.931, 0.979, 0.990). The results demonstrate strong correlations between the reconstructed texture’s mean texture depth (MTD) and friction coefficient (f8) with actual measurements (0.913 and 0.953, respectively), outperforming existing methods. This confirms that the proposed CNN model achieves precise 3D reconstruction of asphalt pavement macrotexture, effectively supporting skid resistance evaluation. To validate engineering applicability, field tests were conducted on pavements with various gradations. The model exhibited excellent robustness under different conditions. Furthermore, based on extensive field data, this study established a quantitative relationship between MTD and friction coefficient, developing a more accurate pavement skid resistance evaluation system to support maintenance decision-making.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Medical Image Segmentation Techniques · Advanced Image Fusion Techniques · Image and Object Detection Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Civil Engineering and Geomatics, Shandong University of Technology, Zibo 255049, China
Funding
the National Natural Science Foundation of China · Natural Science Foundation of Shandong Province
References (30)
Show source-provided references
1. Dong, N., Prozzi, J.A., and Ni, F. (2019). Reconstruction of 3D pavement texture on handling dropouts and spikes using multiple data processing methods. Sensors, 19.10.3390/s19020278 ↗
2. (1997). Characterization of Pavement Texture by Use of Surface Profiles Part 1: Determination of Mean Profile Depth (Standard No. ISO13473-1-1997b).
3. Measurement and modeling of skid resistance of asphalt pavement: A review10.1016/j.conbuildmat.2020.119878 ↗
4. Road Pavement Defect Detection Using High Precision 3D Surveying Technology
5. Research on skid resistance of asphalt pavement based on three-dimensional laser-scanning technology and pressure-sensitive film10.1016/j.conbuildmat.2014.07.015 ↗
6. Evaluation of asphalt pavement texture using multi view stereo reconstruction based on deep learning10.1016/j.conbuildmat.2023.134837 ↗
7. Qi, C., Liu, W., Wu, C., Su, H., and Guibas, L. (2018, January 18–23). Frustum pointnets for 3d object detection from rgb-d data. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.10.1109/CVPR.2018.00102 ↗
8. Shen, H., and Chai, Y. (2007). Summary of Binocular Vision in Computer Vision. Sci. Technol. Inf., 150–151.
9. Schwarz, M., Schulz, H., and Behnke, S. (2015, January 26–30). RGB-D object recognition and pose estimation based on pre-trained convolutional neural network features. Proceedings of the 2015 IEEE International Conference on Robotics and Automation (ICRA), Seattle, WA, USA.10.1109/ICRA.2015.7139363 ↗
10. Imagenet classification with deep convolutional neural networks10.1145/3065386 ↗
11. Long, J., Shelhamer, E., and Darrell, T. (2015, January 7–12). Fully convolutional networks for semantic segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA.10.1109/CVPR.2015.7298965 ↗
12. Deep deconvolution neural network for image super-resolution
13. Faster r-cnn: Towards real-time object detection with region proposal networks10.1109/TPAMI.2016.2577031 ↗
14. Depth estimation from single monocular images using deep hybrid network10.1007/s11042-016-4200-x ↗
15. Laina, I., Rupprecht, C., Belagiannis, V., Tombari, F., and Navab, N. (2016, January 25–28). Deeper depth prediction with fully convolutional residual networks. Proceedings of the 2016 Fourth International Conference on 3D Vision (3DV), Stanford, CA, USA.10.1109/3DV.2016.32 ↗
16. Evaluating asphalt pavement surface texture using 3D digital imaging10.1080/10298436.2018.1483503 ↗
17. Acharya, P.K., and Henderson, T.C. (1988, January 24–29). Parameter estimation and error analysis of range data. Proceedings of the 1988 IEEE International Conference on Robotics and Automation, Philadelphia, PA, USA.
18. A robust method for fitting a plane to point clouds
19. Conditional extremums of functions of multi-variables
20. Ma, F., and Karaman, S. (2018, January 21–25). Sparse-to-dense: Depth prediction from sparse depth samples and a single image. Proceedings of the 2018 IEEE International Conference on Robotics and Automation (ICRA), Brisbane, QLD, Australia.10.1109/ICRA.2018.8460184 ↗
21. Zwald, L., and Lambert, L.S. (2012). The BerHu penalty and the grouped effect. arXiv.
22. Xu, W., Zou, L., Wu, L., and Fu, Z. (2021). Self-Supervised monocular depth learning in low-texture areas. Remote Sens., 13.10.3390/rs13091673 ↗
23. Real-time identification system of asphalt pavement texture based on the close-range photogrammetry10.1016/j.conbuildmat.2019.07.321 ↗
24. Characterizing asphalt pavement 3-D macrotexture using features of co-occurrence matrix
25. Liu, F., Shen, C., and Lin, G. (2015, January 7–12). Deep convolutional neural fields for depth estimation from a single image. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA.10.1109/CVPR.2015.7299152 ↗
26. Wang, P., Shen, X., Lin, Z., Cohen, S., Price, B., and Yuille, A.L. (2015, January 7–12). Towards unified depth and semantic prediction from a single image. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, USA.
27. Hao, Z., Li, Y., You, S., and Lu, F. (2018, January 5–8). Detail preserving depth estimation from a single image using attention guided networks. Proceedings of the 2018 International Conference on 3D Vision (3DV), Verona, Italy.10.1109/3DV.2018.00043 ↗
28. Asphalt pavement macrotexture reconstruction from monocular image based on deep convolutional neural network10.1111/mice.12878 ↗
29. A Comprehensive Survey on Transfer Learning10.1109/JPROC.2020.3004555 ↗
30. (2019). Field Test Methods of Subgrade and Pavement for Highway Engineering (Standard No. JTG 3450-2019).